PCB pattern for René Schmitz "late" MS20 VCF?

fonik · Posted: Mon Dec 10, 2007 5:19 am Post subject:

slightly OT, but i finally decided to post four samples here re the resonance/self oscillation of my wasp clone (edit), interested in your comments.

so1.mp3

Description:

sample #1: selfoscillating, no input signal, 1n4148 diodes in the feedbackpath (i did not normalize the sample to show the difference to sample #2

Download (listen)

Filename:

so1.mp3

Filesize:

90.4 KB

Downloaded:

2021 Time(s)

so2.mp3

Description:

sample #2: same as #1 but with green LEDs.

Download (listen)

Filename:

so2.mp3

Filesize:

110.77 KB

Downloaded:

1853 Time(s)

saw_rm1_2.mp3

Description:

sample #3: saw wave input, resonnance fully turned on, 1n4148 first, then green LEDs.

Download (listen)

Filename:

saw_rm1_2.mp3

Filesize:

551.79 KB

Downloaded:

1996 Time(s)

sq_rm1_2.mp3

Description:

sample #4: same as #3 but with square wave input...

Download (listen)

Filename:

sq_rm1_2.mp3

Filesize:

446.32 KB

Downloaded:

2001 Time(s)

frijitz · Posted: Mon Dec 10, 2007 8:55 pm Post subject:

So -- I have this HP extended to third order. I will have outputs for all three stages, since they all sound different, especially with the anharmonic oscillations. The 3d order has nice sharp resonances when working as a filter.

When using with the distortion/oscillation/drive the 2nd order seems to have more of the anharmonic/chaotic sounds. These seem harder to get in 3d order for some reason, but there are still lots of nice ones.

For clipping I have settled on using back-to-back 7.5V Zeners, which give a nice limiting for the oscillator mode. In addition I will have a switch to add a LED across them for asymmetric distortion.

Tomorrow I start final tweaking, including addition of VC resonance, compensation for the high-frequency Q enhancement and a careful look at DC vs AC coupled operation.

Very Happy

Ian

frijitz · Posted: Tue Dec 11, 2007 4:00 pm Post subject:

v-un-v · Posted: Tue Dec 11, 2007 4:09 pm Post subject:

v-un-v · Posted: Tue Dec 11, 2007 4:11 pm Post subject:

ahh- Wasp clone?? Surprised

Scott Stites · Posted: Tue Dec 11, 2007 10:34 pm Post subject:

Dang, how did I lose track of this thread? Yep, Ian - LP mode will do the same thing. You can have a fair amount of resonance going, remove the signal, and the filter will be self oscillating.

I gotta download Fonik's examples - I've never heard a Wasp filter in the flesh.

Cheerio,
Scott

frijitz · Posted: Wed Dec 12, 2007 8:31 am Post subject:

crashlander42 · Joined: Oct 21, 2006 Posts: 292 Location: Orlando, FL (US)

I just breadboarded the MS20 filter from the schematic on Rene Schmitz's page and I found it very pleasing. Lots of nice squelchy resonance. Twisted Evil

I think I'm going to assemble it in a permanent state tomorrow and toss it in my modular.

Earlier in this thread Ian mentions that the green LED's give it a "dirty" filtering action. I was curious (still learning this electronics magic) what propriety of the green diodes gives it that dirty filtering action, and why green specifically?

Many thanks to whoever feels like explaining that.

frijitz · Posted: Thu Dec 13, 2007 5:27 am Post subject:

frijitz · Posted: Thu Dec 13, 2007 5:34 am Post subject:

I'm still confused a bit about levels. Can anybody tell me what the signal level is in the MS20 at the filter itself? And where on the knob does the oscillation start? In other words, how far beyond the onset of oscillation does the resonance control go.

Thanks!

Very Happy

Ian

Luka · Joined: Jun 29, 2007 Posts: 1003 Location: Melb.

maybe download a "sneaky trial"* version of the korg legacy pack and have a go on the ms20 vst, it would be pretty close wouldnt it?

*not that i condone sneaky trial versions Wink

i own all my software btw
geez im getting all defensive about it hehe

crashlander42 · Joined: Oct 21, 2006 Posts: 292 Location: Orlando, FL (US)

frijitz · Posted: Thu Dec 13, 2007 10:04 pm Post subject:

Scott Stites · Posted: Fri Dec 14, 2007 12:02 pm Post subject:

So...I'm just as curious - can anybody out there answer Ian's question about the MS-20 filters? Tom - you still have an MS-20, don't you?

frijitz · Posted: Sat Dec 15, 2007 6:48 pm Post subject:

frijitz · Posted: Sat Dec 15, 2007 7:45 pm Post subject:

Ah, found the spec sheet. Very Happy

http://www.korganalogue.net/korgms/images/service/ms20/specs.gif
Filter input is 3Vpp max. I.e., max amplitude of 1.5V. This is just as I expected. So the original Korg design looks very tame and ordinary. With Rene's version you not only have 3x the feedback gain, you have 4x the signal amplitude to feed back. Surprised

Ian

Scott Stites · Posted: Sat Dec 15, 2007 8:55 pm Post subject:

Rene is always good for a surprise here and there. Very Happy

Weren't the green LEDs his way of saving on putting three diodes in series X 2, or does it go deeper than that? (IE, the specific voltage drop of green LEDs is what he was after).

frijitz · Posted: Sat Dec 15, 2007 9:59 pm Post subject:

Scott Stites · Posted: Sat Dec 15, 2007 10:31 pm Post subject:

Oh, yes, I was definitely using understatement there. Very Happy

He's an exceedingly intelligent guy, and a very kind person to boot. Heck, a good chunk of my built synthesizer is his stuff.

Really, this filter is all Rene, it came along long before I even knew him or even about him (I've just built and played with it). I think Allison Project has built one, and I'm sure there are others out there. Have you seen Dr. Joerg Schmitz's (no relation) page on it?

Back in January 2004 I visited Rene while on a business trip to Germany, he had so much cool stuff going on, it was almost mind-numbing. I remember he'd built a tube based phase shifter that was just the smoothest thing you'd ever hear. His synth was really interesting too - he'd built the panels out of some kind of blue plastic material - it looked really sharp.

Anyway, on that visit, I traded some parts for some of the MS-20 filter PCBs he'd had made up. I gave a couple of them to Jeff Pontius and kept one (it's in my module, the other one in the module is on protoboard).

Earlier, I was lucky enough to get three VCO3 boards he'd come up with, and those are the VCOs that are mainly heard in my samples. VCO3 has very wide range, which is an understatement. I coined the measurement term "coffebreak" (CB) in order to measure the thing at its lowest setting - I think it was somewhere around 1.5 CB/cycle Laughing

. Earlier this fall, I was driving the Klee up to audio range with it, and when I looked at the frequency I was putting out, it was around 45 kHz, and it still had some room to go up. It's a real workhorse. The KTY part he used for temp compensation is a bit hard to come by on this side of the pond, though. Or, at least it was back then. It would be neat if some more of either of those PCBs could be made.

Cheerio,
Scott

Scott Stites · Posted: Sun Dec 16, 2007 8:19 am Post subject:

I changed the name of the thread slightly to reflect that this version of the Late MS-20 Filter is René's design - it was around even when I began pestering René with a jillion of my newbie questions. I even got that little accent mark in there, which is something I can never quite remember how to do on the keyboard (I pasted it in there)! Notice it did take me a couple of weeks... Laughing

Cheerio,
Scott

Phazinhead · Joined: Feb 12, 2008 Posts: 4 Location: France

Hi everybody !

I'm new in this forum, I like it very much, there are a lot of intersting stuff to read and to discover Smile

I'm not very skilled in electronics, but I'm doing some DIY and I would like to build some filters now.. and as you can guess, I would like to try an MS20 filter clone. I've searched during long time on the internet for PCB and information, which led me to the MS20 filter version of Mr. Takeda and René Schmitz. René's version seems to be the easiest to build and to use standard european components, but I've still some stupid questions in mind .. so if you can help me Smile

:

-What kind of potentiometer should I use for Frequency and Resonnance control ( value and type) ? (.. I found something that said 47K lin for frequency and 100K lin for reso but I'm not shure it is the right value for René's version)

-What are the references of the transistors to use ? Are there many reference of transistor available ?

Thank a lot for your help and your advice !

crashlander42 · Joined: Oct 21, 2006 Posts: 292 Location: Orlando, FL (US)

I just put the Rene Schmitz version of this filter together on the bread board again because I'm planning on throwing one of these together on veroboard and tossing it in my modular.

For some reason the second time around the max resonance seems lower. It's considerably lower than it was the first time I built it. I remember the first time I built it being pleased with it's squelchyness.

Any idea where I would start poking at it to raise the max q?

sneakthief · Joined: Jul 24, 2006 Posts: 569 Location: Berlin

i built takeda's ms-20 clone and couldn't get it to work Sad

it still sits there on my workbench, taunting me.

bubblechamber · Posted: Sun Mar 09, 2008 11:00 pm Post subject:

Hi Crash--- it's a 47k or 50k res pot. the resistor you'll want to play with for higher Q is the 10k between the out and neg of A3. also refer to the schematic here

http://www.electro-music.com/forum/viewtopic.php?highlight=ms20&t=20668

R16 in this schematic
it's a little different and it might work better for you. I think I got the best results with a 2.7K instead of the 10K. mine starts to self osc at around 90%

crashlander42 · Joined: Oct 21, 2006 Posts: 292 Location: Orlando, FL (US)

Thanks! That helped. I also realized that without an attenuated input it's insanely overdriven and I lose a lot of the resonance. I think Ian mentioned something about that and it went in one ear and out the other.

Embarassed

*edit*

Ya. I completely missed the first half of the thread where they pretty much answered my question. The sad part is that I built this once before. Filter sounds great now.